Re: .length vs. count for string length

2016-04-10 Thread Sean Corfield
On 4/10/16, 2:53 AM, "mattias w" wrote: > With clojure 1.8, we got many of these functions, but not str/length and > str/substring. Because we already have `count` and `subs` in clojure.core Sean Corfield -- (904) 302-SEAN An Architect's View -- http://corfield.org/ "If you're not annoying som

Re: .length vs. count for string length

2016-04-10 Thread mattias w
With clojure 1.8, we got many of these functions, but not str/length and str/substring. What am I missing? /mattias Den fredag 1 november 2013 kl. 19:40:42 UTC+1 skrev Sean Corfield: > > This thread made me run a quick audit of our code and we had about a > dozen calls to .length, a dozen cal

Re: .length vs. count for string length

2013-11-01 Thread Sean Corfield
This thread made me run a quick audit of our code and we had about a dozen calls to .length, a dozen calls to .substring, and a handful of calls to .replace - of which a few were in truly performance sensitive code (doing Unicode-related processing across large strings, so they had lots of other Ja

Re: .length vs. count for string length

2013-11-01 Thread vrakade
On Thursday, October 31, 2013 10:37:33 PM UTC-5, Mikera wrote: > > OTOH, count is much more generic since it can handle arbitrary sequences > etc. Also count doesn't require type hints. You should definitely prefer > count when writing most high level code. > Yes, I'd prefer count in higher le

Re: .length vs. count for string length

2013-10-31 Thread Mikera
On Wednesday, 30 October 2013 18:44:47 UTC+8, Alice wrote: > Which one is preferred? > > .length needs to be type hinted, so more verbose. > The performance penalty of count is negligible in most cases. > Doing a quick micro-benchmark with criterium I get: - .length at 5.5 ns - count at 50.0 ns

Re: .length vs. count for string length

2013-10-31 Thread Luc Prefontaine
I meant collection ... not sequence. Luc P. > Strings are character sequences, count is a better option to stay > within the sequence abstraction. > > Lic P. > > > > count does some type checks, but it's negligible in most cases as I already > > said. len can give a clear intention than cou

Re: .length vs. count for string length

2013-10-30 Thread Luc Prefontaine
Strings are character sequences, count is a better option to stay within the sequence abstraction. Lic P. > count does some type checks, but it's negligible in most cases as I already > said. len can give a clear intention than count somtimes. > > I'm not suggesting that it should be included

Re: .length vs. count for string length

2013-10-30 Thread Alice
count does some type checks, but it's negligible in most cases as I already said. len can give a clear intention than count somtimes. I'm not suggesting that it should be included in clojure.string, but if count is currently not preferred over .length, including it can be a good option. On We

Re: .length vs. count for string length

2013-10-30 Thread Baishampayan Ghose
What'd clojure.string/len do any differently than clojure.core/count? count already provides does the fastest possible thing for strings. ~BG On Wed, Oct 30, 2013 at 4:14 PM, Alice wrote: > Which one is preferred? > > .length needs to be type hinted, so more verbose. > The performance penalty of